ual and relatively automatic act, and where the act has been undertaken with no conscious purpose. In such a case the apparent purposelessness of the act is largely a matter of attention. Once failure attracts attention to the outcome, the potential purpose of the act is at once recognized-the failure, in other words, is in effect a failure of definite expectation. This being so, it is at once attributable to some more or less definite factor in the preceding conduct. For it must be recognized that this conduct, however simple it may seem if regarded as a mere objective act (for instance the throwing of the ball at a critical point in a baseball game), is as a piece of conduct exceedingly complex, and capable of many possibilities of modification. Moreover, where conduct is controlled by conceptual thought, it is never directed by a single concept. Just as the import of a concept is expressible only in terms of indirect conditionalities of conduct, so the nature of a given act-its meaning for the individual is expressible only in terms of an organized group of concepts. Thus the modification of the act requisite to satisfy the purpose for which it has been undertaken involves a change. in this initiatory group of concepts, the specific nature of the change demanded depending on the specific nature of the failure in expectation.

A further advantage of the general concept in the control of conduct is to be found in its greater communicability as compared with the concept of the simple object. It is notorious that the development of language, other than that merely expressive of emotion, proceeds pari passu with the growth of general concepts. Imagine the futility of attempting to communicate the meaning of an unclassified, unindividualized object, or the paucity of a language made up wholly of proper names and interjections. Such a state of affairs temporarily exists in every child's life, when it is just beginning to talk. But obviously where speech has progressed no further than the mere attaching of names to different objects there can be little communication of meaning. What makes possible an effective communication is an apprecia

ble degree of organization and mutual dependence of concepts. It is largely the fact that in the general concept the relationship to other concepts has come to be so distinct and to form so prominent a part of the content, that causes the development of conceptual thought and the development of language to coincide. Moreover, it is worth observing that the more indirect a reference to conduct concepts bear, the wider is their range of communicability. That is to say, where such reference is comparatively indirect, communication is possible between individuals whose experiences are comparatively dissimilar; and on the contrary, where concepts are comparatively simple, and refer more directly to conduct, communication is limited to individuals whose habitual daily experiences differ little. We may see this illustrated on a large scale if we observe the effect of national and racial differences upon the cosmopolitan unity of the sciences. In the case of the more abstract sciences, such as logic and mathematics, these differences count for practically nothing. When we come to more concrete sciences, ethics and politics for example, many of the more fruitful developments have had fixed national boundaries.

In this connection it may be well to mention the reciprocal dependence of thought upon language. Whether or not it is possible for general concepts to be formed in the absence of language, we need not attempt to decide; but certainly it must be admitted that no great development of conceptual thought can take place without the aid of language. In general the advantages of language for the formation of general concepts are of two sorts. First, there is the important and evident fact, that it is the instrument of communication. It would be difficult to exaggerate the influence of social intercourse in facilitating the formation and development of general concepts. It at once fixes and corrects old concepts and suggests new ones. We have but to reflect that all science, literature, and art are social products, to realize the part played by social intercourse in our cognitive life. Secondly, the mere fact that a specific verbal

symbol comes to be attached to a given concept operates powerfully in rendering the concept fixed and definite. What frequently happens is that some image of the word (visual, auditory, or motor) becomes the habitual psychological representative through which the concept functions. The word serves thus as a sort of nucleus for the larger and looser associative organization, and thus furthers the conservation of the concept.

No modern treatment of the nature of the general concept can' afford to neglect Berkeley's epoch-making theory. Nowhere in Berkeley's writings is there to be found a more brilliant or effective application of his new-found introspective method than his treatment of abstract ideas. Not only did this treatment revolutionize contemporary theories, but it has exerted a profound influence on the whole later development of psychology.

The gist of Berkeley's account of the matter is contained in the following passage. "Now, if we will annex a meaning to our words, and speak only of what we can conceive, I believe we shall acknowledge that an idea which, considered in itself, is particular, becomes general by being made to represent or stand for all other particular ideas of the same sort. To make this plain by an example, suppose a geometrician is demonstrating the method of cutting a line in two equal parts. He draws, for instance, a black line of an inch in length: this, which in itself is a particular line, is nevertheless with regard to its signification general, since, as it is there used, it represents all particular lines whatsoever; so that what is demonstrated of it is demonstrated of all lines, or, in other words, of a line in general." In a closely succeeding passage, Berkeley notes that a particular idea which acquires the function of standing for a class of ideas undergoes structural modification to this extent: that those features which it has in common with the other members of the class are emphasized or alone attended to, although its peculiar features cannot be wholly banished from consciousness.2

1Principles of Human Knowledge, Introduction, § 12. 21b., § 16..

Berkeley's theory is further amplified by Hume upon two important points; first, with respect to the part played by language in making possible the function of general ideas, and, secondly, with respect to the part played by 'custom' or association in the function of representation. Upon the latter point, Hume remarks that representation implies a certain subdued tendency to revival, such that where the representative idea is used in a connection in which its peculiarities make it no longer typical of the class, the tendency shows itself by the replacement of the unfit representative by a more appropriate member of the class. With regard to the part played by language, Hume commits himself to the extreme view, that it is absolutely essential. The association is not so much between the various ideas of the class, as between each idea and the identical term which denotes them all. Making allowance for this exaggeration, we must acknowledge that Hume strengthens the Berkeleyan theory in no small degree. On the other hand, he fails to notice the structural modification of the representative idea to which Berekley calls attention—most clearly, we may observe, in the latest (1734) edition of the Principles, which Hume may easily not have seen while working upon this part of his Treatise.

Regarding the mutually complementary theories of Berkeley and Hume as substantially one, we find ourselves in fundamental agreement upon the following points: that ideas connected by a relation of resemblance, reinforced by association with a common term, may form a more or less closely unified organization, such that the presence of one of these ideas (or even of the term alone) in consciousness may be accompanied by nascent tendencies to revival of the others; and that it is this complex phenomenon which is referred to under the name of 'general ideas' or 'concepts.'

Our leading divergencies from their view may be summarily expressed as follows:

I. The general concept is not identified with the representative idea, but with the total organization.

2. The point of departure in the formation of the general concept is not in mere ideas but in concepts of objects.

3. The resemblance which forms the bond of association is not (generally speaking) between the ideas themselves, but between the objects denoted by the general concept; and it is fundamentally based upon similarity of import.

4. The resemblance is such as to call for identical behavior in characteristic situations; for it is this necessity for the uniformity of conduct (in spite of individual differences) which fixes attention upon the resemblance and conditions the association based upon it.

Thus far in our discussion of the general concept we have chiefly concerned ourselves with earlier and simpler forms, in order to discover the common characteristics of this type of cognitive organization and its general function in the control of conduct. We now wish to turn our attention to some of the characteristic modifications which the general concept undergoes in the later and more complex stages of mental evolution. These modifications are immediately dependent on what we have tried to exhibit as the most no able feature of the development of cognition, namely the increasing indirectness of its control of conduct.

It will be recalled that in a former chapter pragmatists were criticized as falling into a certain confusion in regard to the end of conduct. The point was made, that while survival is the primary end (in the sense that it is the essential condition for the continuance of conduct), nevertheless it is equally true that happiness also functions as an end in the same sense; and that, moreover, happiness has come to be relatively independent, and much more direct in its influence on the development of conduct. It was further pointed out that what is a common phenomenon of all sorts of activities is to be observed in connection with theoretical activity, namely, that it comes to function in relative independence of its original end. We can now see more clearly why this must be so. As cognition grows more